Virginia Evictions
Richmond & Charlottesville Regions
2018 - Q1 2021

Team members

• Barbara Wilson ◍
  
• Benjamin Teresa ◬
  
• Connor White ◬
• Hannah Woehrle ◬
• Jacob Goldstein-Greenwood ◍ (primary code and document author; contact with code questions)
• Kathryn Howell ◬
  
• Michael Salgueiro ◍
  
• Michele Claibourn ◍

◬ - RVA Eviction Lab
◍ - UVa Equity Center

Last revised: 2021-09-09

Below, we provide some preliminary representations of eviction filings and judgments in Virginia General District Courts for the Richmond and Charlottesville regions, alongside additional characteristics of the locality or ZIP Code in which the court filings occurred. Building off of the work of the RVA Eviction Lab, we are seeking to make a set of eviction and housing justice data readily available in online and interactive formats. This initial presentation is intended (1) to provide a sense of the currently available data and (2) to spark ideas about what information and data would be most useful to include—and in what forms (visual distributions, relations, tables, etc.) and with what capabilities (searching, sorting, selecting, etc.) that material should be presented.

---

The table and visualizations that follow reflect eviction (unlawful detainer) cases filed against residential defendants in the Richmond and Charlottesville regions between 2018 and Q1 2021. For the purposes of this document, the Richmond region comprises Chesterfield County, Henrico County, Hopewell City, Petersburg City, and Richmond City; the Charlottesville region comprises Albemarle County, Charlottesville City, Fluvanna County, Greene County, Louisa County, and Nelson County. (These regional compositions are not unchangeable.)

We begin with raw case data scraped from general district court websites, and we apply a series of cleaning procedures to the data before analysis. Those procedures are described in detail in the drop-down at the bottom of this document. There are a few cleaning procedures that all people viewing these data should be aware of:

• Handling of ZIP Codes: Publicly available unlawful detainer records on court websites do not indicate the exact address under dispute
  • Records list a ZIP and city/county for each defendant and plaintiff in a case
  • We treat the ZIP Code for the first-listed defendant in a given case as the primary ZIP for that case
• Handling of serial cases: We define serial cases as multiple filings by a given plaintiff against a given defendant within a given ZIP Code within a 12-month period
  • When we identify serial cases in the data, we retain the latest in each serial chain
  • This is to emphasize the material outcomes of case filings on defendants: If a plaintiff files three times against a defendant within 12 months, and the first two cases are dismissed on procedural grounds but the third results in a plaintiff judgment, it may be misleading to count the score (from the defendant’s perspective) as “2-1,” given that, ultimately, the defendant was evicted
  • Note: This approach can be adjusted/eliminated depending on the priorities of the advisory committee
• Identification of residential defendants: We apply an in-house name pattern to identify cases filed against residential (v. commercial, governmental, etc.) defendants
  • The table and visualizations below display case data for residential defendants
  • Analyses can include all cases if desired

---

Yearly Evictions and Eviction Rates by Court

2018

2019

2020

Q1 2021

Data Dictionary

• Eviction Rate: Annual eviction rate based on currently available data for each year, # plaintiff judgments / # renting households.
• Cases: Number of unlawful detainer cases filed against residential defendants
  • Note that only one record for duplicate cases or serial cases is included in the totals.
• Plaintiff Judgments: Number of cases ruled in favor of the plaintiff, resulting in an eviction
• Default Judgments: Number of cases ruled in favor of the plaintiff, resulting in an eviction, after a defendant fails to appear
• Rental Prop.: The proportion of households in a spatial unit that are renting (ACS Table B25003)
• Rentals: The number of households in a spatial unit that are renting (ACS Table B25003)
• Median Rent: Median gross rent paid among renting households in a spatial unit (ACS Table B25064)
• Median HH Income: Median household income among all households in a spatial unit (ACS Table B19013)
• Total Pop.: Total population within a spatial unit (ACS Table B02001)
• % POC: Percent of population within a spatial unit who identify as people of color (Black or African American, Hispanic, Asian, American Indian or Alaskan Native, Native Hawaiian or Other Pacific Islander, Multiracial; ACS Table B02001)

---

Eviction Rates by Defendant ZIP Code by Year

2018

2019

2020

Q1 2021

---

Data for Richmond and Charlottesville Regions Across Time

Yearly Trends

Quarterly Trends

---

Click here for additional details on the data cleaning process

We begin with data scraped from Virginia general district court websites by Ben Schoenfeld (virginiacourtdata.org). To these data, we apply a set of cleaning procedures to improve the accuracy of later analyses. Specifically, we do the following:

1. Clean and standardize plaintiff and defendant names

• Court clerk data-entry practices vary, so “SMITH JONES PROPERTIES, L.L.C.” may be entered once as “SMITH-JONES PROPERTIES, LLC” and once as “SMITH JONES PROPERTIES, L.L.C.”; to the best of our abilities, we want to ensure that those cases are identified as being associated with the same entity
• To this end, we:
  • Remove “T/A” and “D/B/A” tags in plaintiff/defendant names
  • Remove dashes, slashes, and periods
  • Remove commas/semicolons that come before a business entity identifier (e.g., “WILLIAMS, LLC”)

2. Extract and clean defendant ZIP Codes

• Not all cases have a ZIP Code listed for the primary defendant that is a valid VA ZIP
  • (We consider ZIPs within the following sets to be valid VA ZIPs: [20100, 20199] and [22000, 24699])
• In the event that a defendant ZIP is not a valid VA ZIP, we convert it to NA so that it does not disrupt by-ZIP tabulations

3. Identify and remove true duplicates (i.e., data entry and/or download errors)
   • When we find multiple cases with the same filing data, judgment, costs, attorney fees, principal value, other awarded amount, primary plaintiff, primary defendant, and defendant ZIP, we remove all but one of the cases
4. Identifying and removing all but the final case in chains of serial cases

• We define serial cases as: Multiple cases in a 12-month period in which plaintiff J filed against defendant K within ZIP Code L
• When we identify chains of serial cases, we retain for analysis the final case in each chain, as it is the most likely case to represenbt the actual outcome of the plaintiff and defendant’s dispute
• For example: Say that the table below includs all cases filed by “ApartCorp” against “Erin Nire” in ZIP Code 12345. Cases of the same color are part of the same chain of serial filings, and bolded cases are retained for analysis

  case_no	pla	def	zip	filed_date
  1	ApartCorp	Erin Nire	12345	01-20-2018
  2	ApartCorp	Erin Nire	12345	03-29-2018
  3	ApartCorp	Erin Nire	12345	04-10-2018
  4	ApartCorp	Erin Nire	12345	01-25-2019
  5	ApartCorp	Erin Nire	12345	03-04-2019

5. Identifying which cases likely have residential vs. non-residential (e.g., commercial, governmental, etc.) defendants

• We use an in-house regular expression pattern to flag cases with non-residential defendants
• The table/visualizations above only reflect data for residential cases

Our full code for cleaning and and summarizing data is available here.